[Bug target/123172] explicit __atomic_*() calls on _BitInt(>64) assume the data is 16-byte aligned

[jakub at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123172

--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
IMHO no, you don't want CMPXCHG16B being used on _Atomic _BitInt(128) and locks
on _BitInt(128), that will mean on the same object some access might use the
former and some the latter.